Fig 1.
Trends comparing efficacy of Cry1Ab Bt sweet corn (Attribute) with non-Bt isogenic hybrids for control of Helicoverpa zea during 1996–2016 in Maryland, USA.
Over years, Bt hybrids (blue circles and shades) had significant increases in the percentage of H. zea damaged ears (A), kernel consumption (B), instar stage (C) and proportion of late instars (D) than non-Bt hybrids (yellow squares and shades). Points represent the raw data, black lines represent the predictions from LMMs, while shaded region denote the upper and lower confidence levels (95% CI).
Fig 2.
Trends comparing efficacy of Cry1A.105+Cry2Ab2 Bt sweet corn (Performance Series) with non-Bt isogenic hybrids for control of Helicoverpa zea during 2010–2016 in Maryland, USA.
Over years, Bt hybrids (blue circles and shades) had significant increases in the percentage of H. zea damaged ears (A), kernel consumption (B), instar stage (C) and proportion of late instars (D) than non-Bt hybrids (yellow squares and shades). Points represent the raw data, black lines represent the predictions from LMMs / GLMM, while shaded region denote the upper and lower confidence levels (95% CI).
Fig 3.
Comparison of weight between susceptible laboratory and field collected Maryland populations of Helicoverpa zea across concentrations of Cry1Ab expressing green leaf tissue.
At every diet concentration other than control (dosage = 0), weight of larvae (mg) after 1 week of feeding assays was significantly higher for the Maryland strain (blue), than the susceptible (orange) strain. Points represent the raw data values; while the squares represent quasi-Poisson GLM estimated means, and the vertical lines denoting the confidence intervals around the mean. Asterisks indicate statistically significant differences (P < 0.05) based on post-hoc comparisons with a Bonferroni correction.
Fig 4.
Mean nightly captures of Helicoverpa zea in black light traps during 1996–2014 in Maryland, USA and its association with damage to Bt corn hybrids.
The population abundance of H. zea at or near study sites decreased over the past two decades (A), and showed no relationship with damage levels in Bt sweet corn hybrids (B). Lines represent predictions from LMMs and shades depict upper and lower confidence intervals (95% CI).
Fig 5.
Results comparing control efficacy variables among Bt and non-Bt hybrid varieties in Maryland, USA in 2008.
The mean damaged ears (A), kernel area consumed, (B), and instars of susceptible larvae (C) was broadly similar among the Cry1Ab expressing Bt hybrids (blue bars), and significantly different (Tukey’s HSD; α = 0.05; denoted by unique alphabets) from non-Bt control hybrids (yellow bar). Black lines represent the mean values and the height of bars above and below the black lines respectively indicate the upper and lower confidence levels (95% CI) estimated through LMMs.
Fig 6.
Results comparing control efficacy in relation to tissue toxicity among Bt and non-Bt sweet corn varieties from Maryland, USA in 2008.
The mean weight, growth inhibition and instars of susceptible larvae feeding on different sweet corn tissues was broadly similar among the Cry1Ab expressing Bt hybrids (blue bars), and significantly different (Tukey’s HSD; α = 0.05) from non-Bt control hybrids (yellow bar). Black lines represent the mean values and height of bars indicate the upper and lower confidence levels (95%CI) estimated through Gaussian GLMs or ANOVA.